Welcome to part 1 of this session. Our goal is to setup Python so that we can get started with Python coding.
Environment setup has traditionally been a major pain-point for Python beginners. Nowadays there are tools that make setting up much easier, and allow you to start writing code with just a few clicks, regardless of which Operating System and hardware you are using. One such tool, Anaconda, is particularly popular among data scientists.
Anaconda is an open-source Python distribution platform. Besides providing the Python executable itself, Anaconda also includes Conda - an open-source source package manager and environment manager for python projects. Conda makes it much easier to build and manage Python projects for scientific research. For users less accustomed to the command-line, Anaconda also provides a graphical user interface for conda, named Anaconda Navigator. We will primarily use Anaconda Navigator throughout this course (though using conda via the console is fine too).
In this section you will:
- Install Anaconda (if using you're using your own computer)
- Open Anaconda Navigator
- Import our Anaconda environment
- Load the environment
- Non-interactively (terminal + text editor)
- Interactively (using IPython)
- School computer - If you're on a school computer, then Anaconda should be available .
- Personal computer - If you're using you're personal computer, download the installer and follow the instructions on screen.
Once Anaconda has been successfully installed, open Anaconda Navigator. It should look like this:
There are two main Navigator tabs you will primarily interact with:
- The Home tab, used to open Applications available within your Environment.
- The Environment tab, lists your python environments and allows you to create and import new environments, or modify existing environments (namely by installing a new package or application).
Learn more about anaconda here.
After you've got accustomed to Anaconda Navigator, you will import an existing Anaconda environment
containing all the Python packages that will be used throughout this course. The environment is
specified as a yaml file
here
(you can copy the
contents
of this file onto a new file in your computer). Use the Import feature of Anaconda Navigator
Environments to import this environment. You may name the environment however you wish, though I
will hereafter refer to it as my-python so I recommend you naming it this way.
Python (virtual) environments are a toll with which to maintain isolated and self-contained python installations and packages, so that requirements and dependencies from different projects do not conflict with each other and wreck havoc across the system. Environments are designed to be easily moved across machines, making your work replicable regardless of the underlying machine details. Read more about:
- Environments
- Python packages - a collection of reusable Python code, structured in modules. A unit of (reusable) code is called a module.
Once the environment has been successfully imported, it will be listed alongside other environments and you may activate it by clicking on it.
Throughout this course, you will run Python in two distinct ways:
- Non-interactively (terminal + text editor)
- Interactively (IPython console or Jupyter notebooks)
We will now show you how to run Python each of these ways. You will have the opportunity to learn how Python works and read more about the different ways to run Python in Part 2 of this weekly course.
The "traditional" way to write and execute Python code follows a workflow similar to:
- open a text editor (a software program that allows you to edit text files), such as Visual Studio Code
- create or open an existing ".py" file (e.g.
my_program.py) - modify its contents by adding new Python code or editing the existing code
- save the contents of the file
- run the python program using the terminal:
python my_program.py,
The terminal is the name for a text-based console
application used to operate a
computer. A terminal prompts you for commands, which you can execute by pressing 'Enter' (return).
You can load your my-python anaconda environment on your "default" terminal application by
selecting "Open with Terminal" when clicking on your environment's "play" button. Note that the
terminal application will depend on the Operating System that you are using.
Once launched, note that the brackets expression (my python) indicates that your my python
environment is activated (and that the "correct" python executable will be called in the terminal).
An easy way to have both the text editor and terminal accessible in one place is to use Visual
Studio Code. You can launch Visual Studio Code from the Home window of Anaconda Navigator. But
first, do not forget to activate your my-python environment! This will make sure that Visual
Studio Code runs in the target environment, as opposed e.g. to the default base environment.
Once you've launched Visual Studio Code successfully, try to:
- Create a file
hello.pywith the line shown above:print("Hello world") - Open a terminal window inside VScode
- Execute your hello world program as shown:
python hello.py
An important thing to keep in mind is that every time you modify your python code, you will have to run it again to see the results. This way of writing Python code contrasts with the method presented next: using interactive Python. Being aware of these two modes of execution is important as both approaches are widely used in Data Science.
The other way to run Python is interactively using IPython. IPython stands for Interactive Python and can allow one to write Python code in a more beginner-friendly way. In interactive mode, the Python program runs continuously and remembers your previous statements. You will learn more about IPython in part 2.
IPython can be executed in two different ways: through a console application (aka terminal, shell)
or Jupyter notebooks (introduced next week). Load your my-python anaconda environment on
IPython shell, using one of two approaches:
- Select "Open with IPython" when clicking on your environment's "play" button. This will open IPython on your default console application.
- Launch the application QtConsole (Home tab of Anaconda Navigator). QtConsole is a more sophisticated IPython console, with additional features. Even though we will not be using its more advanced features, I recommend this approach as it will (otherwise, the default console application will vary from system to system).
Once you've done this, write your first two Python statements, as shown below:
# 1. Print your first message
print("Hello world!")
# 2. Sum two numbers
1 + 2
Can you describe what is happening?
Additional reading:
- Recommended: a very useful resource to learn more about IPython (throughout this course and beyond) is Chapter 0 of the Python Data Science Handbook by Jake VanderPlas.
- IPython website